Michael Hardt, “Affective Labor”
boundary 2 2:26 
Focus on the production of affects in our labor and our social practices has often served as a useful ground for anticapitalist projects, in the context of discourses for instance on desire or on use-value. Affective labor is itself and directly the constitution of communities and collective subjectivities. The productive circuit of affect and value has thus seemed in many respects as an autonomous circuit for the constitutions of subjectivity, alternative to the processes of capitalist valorization. Theoretical frameworks that have brought together Marx and Freud have conceived of affective labor using terms such as desiring production and more significantly numerous feminist investigations analyzing the potentials within what has been designated traditionally as women’s work have grasped affective labor with terms such as kin work and caring labor. Each of these analyses reveal the processes whereby our laboring practices produce collective subjectivities, produce sociality, and ultimately produce society itself.
Such a consideration of affective labor today, however, and this is the primary point of this essay, ought to be situated in the context of the changing role of affective labor in the capitalist economy. In other words, although affective labor has never been entirely outside of capitalist production, the processes of economic postmodernization that have been in course for the past 25 years have positioned affective labor in a role that is not only directly productive of capital but at the very pinnacle of the hierarchy of laboring forms. Affective labor is one face of what I will call “immaterial labor”, which has assumed a dominant position with respect to the other forms of labor in the global capitalist economy. Saying that capital has incorporated and exalted affective labor and that affective labor is one of the highest value-producing forms of labor from the point of view of capital does not mean that, thus contaminated, it is no longer of use to anticapitalist projects. On the contrary, given the role of affective labor as one of the strongest links in the chain of capitalist postmodernization, its potential for subversion and autonomous constitution is all the greater. Within this context we can recognize the biopolitical potential of labor, using biopower here in a sense that both adopts and inverts Foucault’s usage of the term. I want to proceed, then, in three steps: first situating immaterial labor within the contemporary phase of capitalist postmodernization; second, situating affective labor with respect to the other forms of immaterial labor; and finally, exploring the potential of affective labor in terms of biopower.
It has now become common to view the succession of economic paradigms in the dominant capitalist countries since the Middle Ages in three distinct moments, each defined by a privileged sector of the economy: a first paradigm in which agriculture and the extraction of raw materials dominated the economy, a second in which industry and the manufacture of durable goods occupied the privileged position, and the current paradigm in which providing services and manipulating information are at the heart of economic production. The dominant position has thus passed from primary to secondary to tertiary production. Economic modernization named the passage from the first paradigm to the second, from the dominance of agriculture to that of industry. Modernization meant industrialization. We might call the passage from the second paradigm to the third, from the domination of industry to that of services and information, a process of economic postmodernization, or rather, informatization.
The processes of modernization and industrialization transformed and redefined all the elements of the social plane. When agriculture was modernized as industry the farm progressively became a factory, with all of the factory’s discipline, technology, wage-relations, and so forth. More generally, society itself was gradually industrialized even to the point of transforming human relations and human nature. Society became a factory. In the early 20th century Robert Musil reflected beautifully on the transformation of humanity in the passage from the agricultural world to the social factory. “There was a time when people grew naturally into the conditions they found waiting for them and that was a very sound way of becoming oneself. But nowadays, with all this shaking up of things, when everything is becoming detached from the soil it grew in, even where the production of soul is concerned one really ought, as it were, to replace the traditional handicrafts by the sort of intelligence that goes with the machine and the factory.” Humanity and its soul are produced in the very processes of economic production. The processes of becoming human and the nature of the human itself were fundamentally transformed in the qualitative shift of modernization.
In our times, however, modernization has come to an end, or as Robert Kurz says modernization has collapsed. In other words, industrial production is no longer expanding its dominance over other economic forms and social phenomena. A symptom of this shift is manifest in terms of quantitative changes in employment. Whereas the processes of modernization were indicated by a migration of labor from agriculture and mining (the primary sector) to industry (the secondary), the processes of postmodernization or informatization are recognized through the migration from industry to service jobs (the tertiary), a shift that has taken place in the dominant capitalist countries, and particularly in the U.S., since the early 1970s. The terms “services” here covers a large range of activities from health care, education, and finance to transportation, entertainment, and advertising. The jobs for the most part are highly mobile and involve flexible skills. More important, they are characterized in general by the central role played by knowledge, information, communication, and affect. In this sense we can call the postindustrial economy an informational economy.
The claim that the process of modernization is over and that the global economy is today undergoing a process of postmodernization toward an informational economy does not mean that industrial production will be done away with or that it will cease to play an important role, even in the most dominant regions of the globe. Just as the industrial revolution transformed agriculture and made it more productive, so too the informational revolution will transform industry redefining and rejuvenating manufacturing processes—through the integration, for example, of information networks within industrial processes. The new managerial imperative operative here is “treat manufacturing as a service.” In effect, as industries are transformed, the division between manufacturing and services is becoming blurred. Just as through the process of modernization all production became industrialized, so too through the process of postmodernization all production tends toward the production of services, toward becoming informationalized.
The fact that informatization and the shift toward services is most recognizable in the dominant capitalist countries should not lead us back to an understanding of the contemporary global economic situation in terms of stages of development—as if today the dominant countries were informational service economies, their first subordinates industrial economies, and those further subordinated agricultural. For the subordinated countries the collapse of modernization means first of all that industrialization can no longer be seen as the key to economic advancement and competition. Some of the most subordinated regions, such as areas of Sub-Saharan Africa, have been effectively excluded from capital flows and new technologies, from even the illusion of development strategies, and they thus find themselves on the verge of starvation (but we should recognize how postmodernization has imposed this exclusion and nonetheless dominates these regions). Competition for the middle-level positions in the global hierarchy is conducted in large part not through the industrialization but the informatization of production. Large countries with varied economies, such as India, Brazil, and Russia, can support simultaneously all varieties of productive processes: information-based production of services, modern industrial production of goods, and traditional handicraft, agricultural, and mining production. There does not need to be an orderly historical progression among these forms, but rather they mix and coexist; it is not necessary to pass through modernization before informatization—traditional handicraft production can be immediately computerized; cellular phones can be put to use immediately in isolated fishing villages. All of the forms of production exist within the networks of the world market and under the domination of the informational production of services.
The passage toward an informational economy involves necessarily a change in the quality of labor and the nature of laboring processes. This is the most immediate sociological and anthropological implication of the passage of economic paradigms. Information, communication, knowledge, and affect come to play a foundational role in the production process.
A first aspect of this transformation is recognized by many in terms of the change in factory labor—using the auto industry as a central point of reference—from the Fordist model to the Toyotist model. The primary structural change between these models involves the system of communication between the production and consumption of commodities, that is, the passage of information between the factory and the market. The Fordist model constructed a relatively “mute” relationship between production and consumption. The mass production of standardized commodities in the Fordist era could count on an adequate demand and thus had little need to “listen” closely to the market. A feedback circuit from consumption to production did allow changes in the market to spur changes in production but this communication was restricted (due to fixed and compartmentalized channels of planning) and slow (due to the rigidity of the technologies and procedures of mass production).
Toyotism is based on an inversion of the Fordist structure of communication between production and consumption. Ideally, according to this model, the production planning will communicate with markets constantly and immediately. Factories will maintain zero stock and commodities will be produced just in time according to the present demand of the existing markets. This model thus involves not simply a more rapid feedback loop but an inversion of the relationship because, at least in theory, the productive decision actually comes after and in reaction to the market decision. This industrial context provides a first sense in which communication and information have come to play a newly central role in production. One might say that instrumental action and communicative action have become intimately interwoven in informationalized industrial processes. (It would be interesting and useful to consider here how these processes disrupt Habermas’s division between instrumental and communicative action, just as in another sense they do Arendt’s distinctions among labor, work, and action.) One should quickly add, however, that this is an impoverished notion of communication, the mere transmission of market data.
The service sectors of the economy present a richer model of productive communication. Most services indeed are based on the continual exchange of information and knowledges. Since the production of services results in no material and durable good, we might define the labor involved in this production as immaterial labor—that is, labor that produces an immaterial good, such as a service, knowledge, or communication. One face of immaterial labor can be recognized in analogy to the functioning of a computer. The increasingly extensive use of computers has tended progressively to redefine laboring practices and relations (along with indeed all social practices and relations). Familiarity and facility with computer technology is becoming an increasingly general primary qualification for work in the dominant countries. Even when direct contact with computers is not involved the manipulation of symbols and information along the model of computer operation is extremely widespread. One novel aspect of the computer is that it can continually modify its own operation through its use. Even the most rudimentary forms of artificial intelligence allow the computer to expand and perfect its operation based on interaction with its user and its environment. The same kind of continual interactivity characterizes a wide range of contemporary productive activities throughout the economy, whether computer hardware is directly involved or not. In an earlier era workers learned how to act like machines both inside and outside the factory. Today, as general social knowledge becomes ever more a direct force of production, we increasingly think like computers and the interactive model of communication technologies becomes more and more central to our laboring activities. Interactive and cybernetic machines become a new prosthesis integrated into our bodies and minds and a lens through which to redefine our bodies and minds themselves.
Robert Reich calls this type of immaterial labor “symbolic-analytical services”—tasks that involve “problem-solving, problem-identifying, and strategic brokering activities.” This type of labor claims the highest value and thus Reich identifies it as the key to competition in the new global economy. He recognizes, however, that the growth of these knowledge-based jobs of creative symbolic manipulation implies a corresponding growth of low value and low skill jobs of routine symbol manipulation, such as data entry and word processing. Here begins to emerge a fundamental division of labor within the realm of immaterial processes.
The model of the computer, however, can only account for one face of the communicational and immaterial labor involved in the production of services. The other face of immaterial labor is the affective labor of human contact and interaction. This is the aspect of immaterial labor that economists like Reich are less likely to talk about, but that seems to me the more important aspect, the binding element. Health services, for example, rely centrally on caring and affective labor, and the entertainment industry and the various culture industries are likewise focussed on the creation and manipulation of affects. To one degree or another this affective labor plays a certain role throughout the service industries, from fast food servers to providers of financial services, embedded in the moments of human interaction and communication. This labor is immaterial, even if it is corporeal and affective, in the sense that its products are intangible: a feeling of ease, well-being, satisfaction, excitement, passion—even a sense of connectedness or community. Categories such as in-person services or services of proximity are often used to identify this kind of labor, but what is essential to it, its “in-person” aspect, is really the creation and manipulation of affects. Such affective production, exchange, and communication is generally associated with human contact, with the actual presence of another, but that contact can be either actual or virtual. In production of affects in the entertainment industry, for example, the human contact, the presence of others, is principally virtual, but not for that reason any less real.
This second face of immaterial labor, its affective face, extends beyond the model of intelligence and communication defined by the computer. Affective labor is better understood by beginning from what feminist analyses of “women’s work” have called “labor in the bodily mode.” Caring labor is certainly entirely immersed in the corporeal, the somatic, but the affects it produces are nonetheless immaterial. What affective labor produces are social networks, forms of community, biopower.
Here one might recognize once again that the instrumental action of economic production has merged with the communicative action of human relations. In this case, however, communication has not been impoverished but rather production has been enriched to the level of complexity of human interaction. Whereas in a first moment, in the computerization of industry for example, one might say that communicative action, human relations, and culture have been instrumentalized, reified, and “degraded” to the level of economic interactions, one should add quickly that through a reciprocal process, in this second moment, production has become communicative, affective, de-instrumentalized, and “elevated” to the level of human relations—but of course a level of human relations entirely dominated by and internal to capital. (Here the division between economy and culture begins to break down.) In the production and reproduction of affects, in those networks of culture and communication, collective subjectivities are produced and sociality is produced—even if those subjectivities and that sociality are directly exploitable by capital. This is where we can realize the enormous potential in affective labor.
I do not mean to argue that affective labor itself is new or that the fact that affective labor produces value in some sense is new. Feminist analyses in particular have long recognized the social value of caring labor, kin work, nurturing, and maternal activities. What are new, on the other hand, are the extent to which this affective immaterial labor is now directly productive of capital and the extent to which it has become generalized through wide sectors of the economy. In effect, as a component of immaterial labor, affective labor has achieved a dominant position of the highest value in the contemporary informational economy. Where the production of soul is concerned, as Musil might say, we should no longer look to the soil and organic development, nor to the factory and mechanical development, but rather to today’s dominant economic forms, that is, to production defined by a combination of cybernetics and affect.
This immaterial labor is not isolated to a certain population of workers, say computer programmers and nurses, who would form a new potential labor aristocracy. Rather immaterial labor in its various guises (informational, affective, communicative, and cultural) tends toward being spread throughout the entire workforce and throughout all laboring tasks as a component, larger or smaller, of all laboring processes. That said, however, there are certainly numerous divisions within the realm of immaterial labor—international divisions of immaterial labor, gender divisions, racial divisions, and so forth. As Robert Reich says, the U.S. government will strive as much as possible to keep the highest value immaterial labor in the United States and export the low value tasks to other regions. It is a very important task to clarify these divisions of immaterial labor, which I should point out are not the divisions of labor we are used to, particularly with regard to affective labor.
In short, we can distinguish three types of immaterial labor that drive the service sector at the top of the informational economy. The first is involved in an industrial production that has been informationalized and has incorporated communication technologies in a way that transforms the industrial production process itself. Manufacturing is regarded as a service and the material labor of the production of durable goods mixes with and tends toward immaterial labor. Second is the immaterial labor of analytical and symbolic tasks, which itself breaks down into creative and intelligent manipulation on one hand and routine symbolic tasks on the other. Finally, a third type of immaterial labor involves the production and manipulation of affects and requires (virtual or actual) human contact and proximity. These are the three types of labor that drive the postmodernization or informationalization of the global economy.
By biopower I understand the potential of affective labor. Biopower is the power of the creation of life; it is the production of collective subjectivities, sociality, and society itself. The focus on affects and the networks of the production of affects reveals these processes of social constitution. What is created in the networks of affective labor is a form-of-life.
When Foucault discusses biopower he sees it only from above. It is patria potestas, the right of the father over the life and death of his children and servants. More important, biopower is the power of the emerging forces of governmentality to create, manage, and control populations—the power to manage life. Other more recent studies have extended Foucault’s notion, casting biopower as the rule of the sovereign over “naked life,” life distinct from its various social forms. In each case, what is at stake in power is life itself. This political passage towards the contemporary phase of biopower corresponds to the economic passage of capitalist postmodernization in which immaterial labor has been cast in the dominant position. Here too in the creation of value and the production of capital what is central is the production of life, that is, the creation, management, and control of populations. This Foucauldian view of biopower, however, only poses the situation from above, as the prerogative of a sovereign power. When we look at the situation from the perspective of the labor involved in biopolitical production, on the other hand, we can begin to recognize biopower from below.
The first fact we see when we adopt this perspective is that the labor of biopolitical production is strongly configured as gendered labor. Indeed various streams of feminist theory have already provided extensive analyses of the production of biopower from below. A current of ecofeminism, for example, employs the term biopolitics (in a way that might seem at first sight quite different from that of Foucault) to refer to the politics of the various forms of biotechnology that are imposed by transnational corporations on populations and environments, primarily in subordinated regions of the world. The Green Revolution and other technological programs that have been cast as means of capitalist economic development actually have brought with them both devastation for the natural environment and new mechanisms for the subordination of women. These two effects, however, are really one. It is primarily the traditional role of women, these authors point out, to fulfill the tasks of reproduction that have been most severely affected by the ecological and biological interventions. From this perspective, then, women and nature are dominated together but they also work together in a cooperative relationship, against the assault of biopolitical technologies, to produce and reproduce life. Staying alive: politics has become a matter of life itself and the struggle has taken the form of a biopower from above against a biopower from below.
In a very different context numerous feminist authors in the United States have analyzed the primary role of women’s labor in the production and reproduction of life. In particular, the caring labor involved in maternal work (distinguishing maternal work from the biologically specific aspects of birthing labor) has proven to be an extremely rich terrain for the analysis of biopolitical production. Biopolitical production here consists primarily in the labor involved in the creation of life—not the activities of procreation, but the creation of life precisely in the production and reproduction of affects. Here we can recognize clearly how the distinction between production and reproduction breaks down, as does that between economy and culture. Labor works directly on the affects; it produces subjectivity, it produces society, it produces life. Affective labor, in this sense, is ontological—it reveals living labor constituting a form of life and thus demonstrates again the potential of biopolitical production.
We should add immediately, however, that we cannot simply affirm either of these perspectives in an unqualified way, without recognizing the enormous dangers they pose. In the first case, the identification of women and nature risks naturalizing and absolutizing sexual difference, in addition to posing a spontaneous definition of nature itself. In the second case, the celebration of maternal work could easily serve to re-enforce both the gendered division of labor and the familial structures of oedipal subjection and subjectification. Even in these feminist analyses of maternal labor it is clear how difficult it can be at times to dislodge the potential of affective labor from both the patriarchal constructions of reproduction and the subjective black hole of the family. These dangers, however, important though they might be, do not negate the importance of recognizing the potential of labor as biopower, a biopower from below.
This biopolitical context is precisely the ground for an investigation of the productive relationship between affect and value. What we find here is not so much the resistance of what might be called “affectively necessary labor,” but rather the potential of necessary affective labor. On one hand affective labor, the production and reproduction of life, has become firmly embedded as a necessary foundation for capitalist accumulation and patriarchal order. On the other hand, however, the production of affects, subjectivities, and forms of life present an enormous potential for autonomous circuits of valorization, and perhaps for liberation.
 Robert Musil, The Man Without Qualities, vol. 2, trans. Sophie Wilkins (New York: Vintage, 1996) 367
 On the employment shifts in the dominant countries, see Manuel Castells and Yuko Aoyama, “Paths towards the informational society: Employment structure in G-7 countries, 1920-90,” International Labour Review 133:1 (1994): 5-33.
 François Bar, “Information Infrastructure and the Transformation of Manufacturing,” in The New Information Infrastructure: Strategies for U.S. Policy, ed. William Drake (New York: Twentieth Century Fund Press, 1995), 56.
 On the comparison between the Fordist and Toyotist models, see Benjamin Coriat, Penser à l’envers: travail et organisation dans l’entreprise japonaise (Paris: Christian Bourgois, 1994).
 I am thinking primarily of Jürgen Habermas, The Theory of Communicative Action, trans. Thomas McCarthy (Boston: Beacon Press, 1984); and Hannah Arendt, The Human Condition (Chicago: University of Chicago Press, 1958). For an excellent critique of Habermas’s division between communicative and instrumental action in the context of economic postmodernization, see Christian Marazzi, Il posto dei calzini: la svolta linguistica dell’economia e i suoi effetti nella politica (Bellinzona, Switzerland: Casagrande, 1995), 29-34.
 For a definition and analysis of immaterial labor, see Maurizio Lazzarato, “Immaterial Labor,” in Radical Thought in Italy, ed. Paolo Virno and Michael Hardt (Minneapolis: University of Minnesota Press, 1996), 133-47.
 Peter Drucker understands the passage toward immaterial production as the complete destruction of the traditional categories of political economy. “The basic economic resource—‘the means of production,’ to use the economist’s term—is no longer capital, nor natural resources (the economist’s ‘land’), nor ‘labor.’ It is and will be knowledge.” Peter Drucker, Post-Capitalist Society, (New York: Harper, 1993), 8. What Drucker does not understand is that knowledge is not given but produced and that its production involves new kinds of means of production and labor.
 Marx uses the term “general intellect” to refer to this paradigm of productive social activity. “The development of fixed capital indicates to what degree social knowledge has become a direct force of production, and to what degree, hence, the conditions of the process of social life itself have come under the control of the general intellect and been transformed in accordance with it. To what degree the powers of social production have been produced, not only in the form of knowledge, but also as immediate organs of social practice, of the real life process.” Karl Marx, Grundrisse, trans. Martin Nicolaus (New York: Vintage, 1973), 706.
 Robert Reich, The Work of Nations: Preparing Ourselves for 21st-Century Capitalism (New York: Knopf, 1991), 177.
 See Dorothy Smith, The Everyday World as Problematic: A Feminist Sociology (Boston: Northeastern University Press, 1987), 78-88.
 See primarily Michel Foucault, The History of Sexuality, vol. 1, trans. Robert Hurley (New York: Vintage, 1978), 135-45.
 See Giorgio Agamben, Homo Sacer, (Turin: Einaudi, 1995); and “Form-of-Life,” trans. Cesare Casarino, in Radical Thought in Italy, ed. Paolo Virno and Michael Hardt (Minneapolis: University of Minnesota Press, 1996), 151-56.
 See Vandana Shiva and Ingunn Moser, ed., Biopolitics: A Feminist and Ecological Reader (London: Zed Books, 1995); and more generally Vandana Shiva, Staying Alive: Women, Ecology and Survival in India (London: Zed Books, 1988).
 See Sara Ruddick, Maternal Thinking: Towards a Politics of Peace (New York: Ballantine Books, 1989).
 On the ontologically constitutive capacities of labor, specifically in the context of feminist theory, see Kathi Weeks, Constituting Feminist Subjects (Ithaca: Cornell University Press, 1998), 120-51.
 See Gayatri Chakravorty Spivak, “Scattered Speculations on the Question of Value,” in In Other Worlds (New York: Routledge, 1988) 154-75.